)]}'
{
  "commit": "321bcf92163038e2b96fd3bf9bc29f755c81d9ef",
  "tree": "68a1e6b659cac5ffdadb4b7412fa5b7a4244541b",
  "parents": [
    "b68680e4731abbd78863063aaa0dca2a6d8cc723"
  ],
  "author": {
    "name": "J. Bruce Fields",
    "email": "bfields@citi.umich.edu",
    "time": "Sun Oct 21 16:41:38 2007 -0700"
  },
  "committer": {
    "name": "Linus Torvalds",
    "email": "torvalds@woody.linux-foundation.org",
    "time": "Mon Oct 22 08:13:18 2007 -0700"
  },
  "message": "dcache: don\u0027t expose uninitialized memory in /proc/\u003cpid\u003e/fd/\u003cfd\u003e\n\nWell, it\u0027s not especially important that target-\u003ed_iname get the contents\nof dentry-\u003ed_iname, but it\u0027s important that it get initialized with\n*something*, otherwise we\u0027re just exposing some random piece of memory to\nanyone who reads the link at /proc/\u003cpid\u003e/fd/\u003cfd\u003e for the deleted file, when\nit\u0027s still held open by someone.\n\nI\u0027ve run a test program that copies a short (\u003c36 character) name ontop of a\nlong (\u003e\u003d36 character) name and see that the first time I run it, without\nthis patch, I get unpredicatable results out of /proc/\u003cpid\u003e/fd/\u003cfd\u003e.\n\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "2bb3f7ac683b31e55cc783db67a91ad8e480c63a",
      "old_mode": 33188,
      "old_path": "fs/dcache.c",
      "new_id": "d9ca1e5ceb92f02302fdeedb7bc1e2b6e5768f2e",
      "new_mode": 33188,
      "new_path": "fs/dcache.c"
    }
  ]
}